19 Oct Today’s Forecasting Challenges Facing Risk Managers
by: Ian Palao
Senior executives and risk managers know the importance of an accurate long term and short term load forecast. Many of them have significant challenges consistently producing accurate results. The following article outlines the most common reasons why.
1) Miscalculating substantial growth (or attrition).
Adding (or losing) 100’s or even 1,000’s of customers per day can dramatically change the way your portfolio consumes energy. Each individual account is unique in its usage patterns. Trying to keep track of this with a purely ‘top-down’ methodology can make your job an endless moving target. These methods tend to treat customers as a clone of every other customer and therefore cannot accurately forecast in a high-churn environment.
2) You’re using nothing but spreadsheets and your macros/links get hosed.
If, like most companies, you’re using Excel to produce your forecasts, you’ve probably developed a highly complex and extremely fragile set of macros (which eventually break for one reason or another). Reconnecting links and ensuring data integrity is a time-consuming nightmare and is often managed by only one person. Files crash and data gets corrupted. Broken links cannot be maintained or updated automatically. Keeping this delicate jumble of spreadsheets in working order is a significant challenge and takes your eyes off of the ultimate prize — accurate results. Risk managers spend more time on menial tasks instead of spending that time effectively managing their risk.
3) Using utility load profiles for accounts that don’t ‘fit the mold’.
The big manufacturing facility you just gained doesn’t have the usage characteristics of grandma’s house down the street, nor does it look like that big box store downtown. Simply relying on the power curves and profiles provided by the utility can cost thousands of dollars in a single day. Depending on your portfolio mix in a particular zone, a single account can throw off your entire forecast, making the entire zone extremely unprofitable.
4) Complete lack of faith in the forecast you’re seeing.
After all is said and done, your current methods are lacking due to various reasons. This leaves you wondering if you can really trust your forecast. Have you tested more than one forecasting model? Have you taken into account meter changes, cancel/re-bills, updated weather forecasts etc? Do you have ‘what-if’ scenarios for more advanced seasonal strategies? If you have limited weather data and/or the inability to run more advanced analytics, you may be leaving your forecast(s) to chance. Not having all of the necessary data at your fingertips exactly when you need it leaves you with no choice but to rely on the old ‘thumb-in-the-air’ methodology that many risk managers tend to fall back on.
5) Relying on last week’s forecast for an upcoming extreme weather event.
If an arctic blast (or heat wave) is forecast for next week, relying on last week’s forecast isn’t the best method for addressing the extreme weather patterns coming to a market near you. Using ‘normal’ weather for load forecasting when the weather is anything but ‘normal’ can lead to disaster. Most methods and/or manual techniques cannot process data quickly enough to accurately model energy usage. This lack of weather-responsiveness often leaves risk managers guessing. These extreme events often carry extremely high price tags due to increased commodity prices.
6) Assuming staff will never call in sick or resign.
Where would your company be if your main risk manager got injured, sick or resigned? Would the trains keep moving in a smooth and timely manner? Relying on the manual forecasting methods with limited key personnel is risky and is not sustainable due to the increasing complexity and real-time nature of the energy industry. Individual people become critical cogs in the operation. When a staff member is absent, risk and supply managers are often left scrambling to simply complete basic tasks when they should be focused on the bigger picture. Manual labor replaces critical thought which can dramatically jeopardize profit margins.
If any of these challenges sound familiar, you’re not alone. However, there are several solutions available on the market to help reduce or completely eliminate some (or all) of these challenges.
Ian Palao of EscoWare